From Steam to Diesel: Managerial Customs and Organizational Capabilities in the Twentieth-Century American Locomotive Industry

From steam to diesel

Princeton Studies in Business and Technology

Series Editor

David Hounshell

Before the Computer: Mechanical Information Processing in the United States, 1865–1956, James W. Cortada

Politics and Industrialization: Early Railroads in the United States and Prussia, Colleen A. Dunlavy

From Steam to Diesel

Managerial Customs and Organizational Capabilities in the Twentieth-Century American Locomotive Industry

Albert J. Churella

Princeton University Press Princeton, New Jersey

Copyright © 1998 by Princeton University Press

Published by Princeton University Press, 41 William Street,

Princeton, New Jersey 08540

In the United Kingdom: Princeton University Press,

Chichester, West Sussex

All Rights Reserved

Library of Congress Cataloging-in-Publication Data

Churella, Albert J.

From steam to diesel: managerial customs and organization capabilities in the

twentieth-century American locomotive industry / Albert J. Churella.

p.    cm.

Includes bibliographical references and index.

eISBN 1-4008-0120-6

This book has been composed in Caledonia

http://pup.princeton.edu

Contents

Acknowledgments

Introduction

I.Steam vs. Diesel: Capabilities and Requirements of a Radically New Technology

II.Internal-Combustion Railcars: Springboard to Participation in the Diesel Locomotive Industry

III.First-Mover Advantages and the Decentralized Corporation

IV.ALCo and Baldwin: Established Companies, New Technologies

V.Policy and Production during World War II

VI.Postwar Dieselization and Industry Shakeout

VII.The Era of Oligopoly

Conclusion

Notes

Bibliography

Index

Acknowledgments

As this book is a revision of my doctoral dissertation, I am deeply indebted to my adviser, Mansel Blackford. Without his insightful guidance and patient understanding, my interest in this project would never have come to fruition. K. Austin Kerr and William Childs have also given generously of their time and advice in seminars and through individual consultation. I appreciate greatly the comments and suggestions made by David Hounshell and Steven Usselman. Their insights have helped me to see the shortcomings in my original dissertation as I have worked to transform it into a book. I have had the opportunity to develop my study of the locomotive industry in a recent article in the Business History Review, and I am grateful to that journal for permitting me to reprint portions of that article in this work. Conferences of the Business History Society, the Society for the History of Technology, the Economic and Business Historical Society, and the Ohio Academy of History have also provided a forum for my work during its formative stages.

This book has absorbed many hours of archival research. Unfortunately, since General Motors and General Electric are still actively competing in the diesel locomotive industry, their records are not open to scholars. Furthermore, when the old steam locomotive builders left the industry, railroad enthusiasts and historians often saved a treasure trove of locomotive photographs, drawings, and blueprints, but could not preserve correspondence, memoranda, reports, or other written sources. To the extent that they were able to do so, I am eternally grateful. In particular, I would like to thank railroad historians and experts Jim Mischke and Wallace Abbey for giving me expert technical advice on the locomotive industry and its products.

The research for this book was made possible by the generous assistance of librarians, archivists, and scholars throughout the United States. Those who have helped include Tab Lewis and James Cassedy at the National Archives in Washington; Gregory J. Plunges at the National Archives, Northeast Region; Beverly Watkins at the National Archives, Great Lakes Region; Elaine Fidler and Ron Wiercioch at the U.S. Department of Justice; Joyce Koeneman and Andrea Cheney at the Association of American Railroads Library; Mark Cedeck, Curator of the John W. Barriger III Collection at the St. Louis Mercantile Association Library; Charles Bates and John Keller at the Allen County (Ohio) Historical Society; Jim Bachorz at the American Locomotive Historical Society; Sharon Nelson at the Pennsylvania State Archives; George Deeming at the Railroad Museum of Pennsylvania; Connie Menninger at the Kansas State Historical Society; Denise Conklin at the Pattee Library of the University of Pennsylvania; Christopher Baer, Lynn Catanese, and Marjorie McNinch at the Hagley Museum and Library; the staff of the George Arents Research Library at Syracuse University; the staff of the Minnesota Historical Society in St. Paul; Richard Scharchburg and Bruce Watson at the General Motors Institute; George Wise at General Electric, L. Dale Patterson at the University of Louisville; Bridget Burke at the Smithsonian Institution; Blaine Lamb at the California State Railroad Museum; Kay Bost, Curator of the DeGolyer Library at Southern Methodist University; Reese Jenkins, editor of the Thomas A. Edison papers at Rutgers University; Roland Marchand at The University of California at Davis; Leonard Reich at Colby College; and Richard Tedlow at Harvard University. All of these individuals have been more than generous in their support of my research and writing.

Research can often be an expensive proposition, especially when extensive travel is involved. Several grants have helped to reduce this financial burden. The Ohio State University Graduate School provided a Graduate Student Alumni Research Award, while the Ohio State University History Department awarded a Ruth Higgins Summer Research Fellowship. Finally, The University of Illinois Foundation generously provided a John E. Rovensky Fellowship in Business and Economic History.

This book would doubtless never have been written without the ongoing support and sympathy of my parents, Albert A. and Helen Churella. My greatest debt—one that I can not adequately express in words—must go to my wife, Marianne Holdzkom. In spite of research trips, writer’s block, and numerous moves, she has exhibited compassion, understanding, and love beyond measure. Finally, while many individuals have contributed generously to my work, any errors or omissions are solely my own responsibility.

From Steam To Diesel

Introduction

For more than a hundred years, from the 1830s through the 1940s, steam locomotives formed the main power source for railroads throughout the world. Beyond their duties as haulers of freight and passenger traffic, steam locomotives symbolized both the romance of the rails and the industrial might of the American economy.¹ For all of their undeniable power and majesty, however, steam locomotives were notoriously inefficient and costly to maintain. For all of their symbolic reference to American industry, steam locomotives remained customized, hand-built products in a nation that enthusiastically embraced mass production. Any form of motive power that could overcome the technological limitations of the steam locomotive would, as a radical technological discontinuity, have profound effects on established steam locomotive producers.

Diesel locomotive technology represented just such a discontinuity. Diesel locomotives not only made more efficient use of fuel than did steam locomotives, they also cost less to service and repair. Beyond that, diesel locomotive technology was much more amenable to mass-production methods and the savings that resulted from these manufacturing techniques. The technological advantages of diesel locomotives had become evident by the mid-1930s, and diesels quickly replaced steam locomotives in most railroad assignments. By the late 1950s, it was difficult to find a steam locomotive in revenue service anywhere in the United States.

This book does not explore the development of diesel locomotive technology, per se. Rather, it explores the ways in which corporations and corporate executives responded—or failed to respond—to technological change. The locomotive industry provides the backdrop for an exploration of the processes of innovation and technological diffusion, the role of managerial culture and tradition in decision-making processes, and the ways in which both random events and exogenous forces helped to shape competitive patterns.

The dieselization revolution that swept through the American railroad industry in the decade following World War II changed forever the structure of the locomotive industry in the United States. Three companies—the American Locomotive Company (ALCo), the Baldwin Locomotive Works, and the Lima Locomotive Works—had long controlled steam locomotive production. Although all three firms built diesel locomotives, they had abandoned locomotive production by the late 1960s. In their place, the Electro-Motive Company (later the Electro-Motive Division of General Motors) and General Electric came to dominate the American diesel locomotive industry.²

The transition from steam to diesel locomotive production thus offers an early example of a complete turnover of firms within a single major industry. As a part of this turnover, the size and profitability of both ALCo and Baldwin plummeted. In 1917, the two companies were ranked at #52 and #62, respectively, among the two hundred largest industrial corporations in the United States. By 1948, their rankings had fallen to #145 and #143, and continued to decline in the years that followed. With only two exceptions (Great Western Sugar and Willys-Overland), no industrial corporations in the history of American business fell so far, so fast.³ At the same time, two large diversified corporations came to dominate the diesel locomotive industry.⁴

Context and Scope of the Dieselization Story

The locomotive industry during the twentieth century provides a fascinating case study of technological change and industrial transformation precisely because this total firm turnover existed. The speed and completeness with which the Schumpeterian gales of creative destruction swept through the locomotive industry raises several important issues.⁵

The first set of questions concerns the source of technological change, diffusion, and progress. Technological discontinuities may arise from a variety of sources, such as the organization, a technological system, or a community of technological practitioners.⁶ Historians and economists have often been reluctant to explore the root causes of the technological changes that in turn contribute to alterations in firm behavior.⁷ The evolution of the locomotive industry provides insights into this often unexplored black box of technology.⁸ Furthermore, activities inside the black box frequently occur in a random, chancy, and often irrational manner.⁹

Inasmuch as the diesel engine succeeded in railroad service only half a century after its initial development, after first making a wide detour through the realm of maritime motive-power technology, the issue of technology transfer is of considerable importance to this case study. While lone individuals—the heroic inventors—played an important role in the development and diffusion of diesel locomotive technology, communities of technical practitioners were also able to impose their values on corporate hierarchies, at least during the early stages of the industry.¹⁰ In particular, the locomotive industry indicates that the role of the heroic inventor (or, at least, the zealous promoter) may have been underemphasized in technological histories of large corporate entities.¹¹

The history of the locomotive industry is also the story of innovation in both technology and business practices. Since other historians have explored the invention of the diesel engine and since the subject is beyond the scope of this volume, innovation assumes greater weight in this study.¹² While success in the locomotive industry at first depended largely on product innovation, process innovation—the development of new operating routines, particularly in the realm of production controls and marketing initiatives—ultimately held more importance.¹³

Diesel locomotive technology, by its very nature, determined production methods and competitive patterns, suggesting that soft determinism and related historical paradigms such as technological momentum and path dependency had considerable importance in the locomotive industry. Conversely, stochastic elements, such as historical chance and luck, played a considerable role in the development of the locomotive industry and in the diffusion of diesel locomotive technology. Exogenous—and largely coincidental—technological linkages also affected the pace of technological change in the locomotive industry.¹⁴

This study also addresses the role of patents and proprietary information as either an inducement or a barrier to technological diffusion. Patents, in the end, were of scant importance in the locomotive industry, and ALCo, Baldwin, and Lima suffered few setbacks because they lacked access to proprietary information. At the same time, companies were often unable to guard against leakages from their stock of proprietary knowledge.¹⁵ The locomotive industry thus differed from the American railroad industry, in which patent pools encouraged technological diffusion while doing little to protect the proprietary rights of the inventor.¹⁶

This work also explores the relative effectiveness of varied corporate responses to technological change. American locomotive builders employed a number of competitive strategies in response to the change in technology from steam to diesel motive power. These included joint ventures, such as that between ALCo and General Electric, horizontal combination, as in a merger between Lima and Baldwin, and vertical integration, in the case of General Motors. These varied organizational responses to technological change, in turn, relate to the larger issue of how established companies adapt, or fail to adapt, to technological, political, social, and economic changes.

The locomotive industry also offers the possibility of assessing the potential for large diversified firms to exploit technological innovation and thereby fulfill larger social and economic needs.¹⁷ The financial and technical resources of GM and GE facilitated their capture of the diesel locomotive market, indicating that modern diversified firms may be particularly efficient at exploiting technological innovations. Companies often have naturally differing abilities to respond to technological change, and such asymmetries—whether based on firm size or on other factors—can cause profound changes in competitive patterns and market structures.¹⁸ While the expansion of firms into related product lines may initially increase competition, this process may lead ultimately to a reduction in competition; and this, in turn, may produce economically or socially undesirable consequences. In particular, this work examines GM’s effects on the structure of the locomotive industry in light of the Justice Department’s intense scrutiny of that corporation during the 1950s and 1960s. A related issue concerns the degree to which firms—large or small—can exert control over their technology and their markets.¹⁹

While the locomotive industry never assumed such a high profile as, for example, automobiles or aerospace, the importance of railroads to the American economy, particularly in wartime, and the emergence of a large and potentially monopolistic firm in the locomotive industry ensured that government policy would intersect with technological change and business strategy. As such, the locomotive industry offers insights into the effects of government policy, both proactively and as an attempt to alter established competitive patterns.²⁰

In recent years, an ongoing dialogue has developed between historians who examine mass production, such as Alfred D. Chandler Jr., and those who have turned their attention to smaller, more specialized producers of small-batch or custom-manufactured products.²¹ This work will describe how the locomotive industry, between 1930 and 1960, evolved from flexible small-batch production methods to manufacturing techniques that more closely resembled mass production. In other words, ALCo, Baldwin, and Lima had to respond to a radical technological discontinuity by making equally radical changes in their style of production. Managers at all three companies proved understandably reluctant to abandon time-tested methods of small-batch production, yet, at the same time, they chose to remain in the locomotive industry, even though this industry increasingly devalued small-batch manufacturing techniques.

The ultimate inability of the three established steam locomotive producers to make the transition from small-batch custom production to standardized near-mass production in turn raises a larger issue; namely, the extraordinary difficulty of transforming firm routines to accommodate radical technological discontinuities. In the case of the locomotive industry, ALCo, Lima, and Baldwin mastered the art of making incremental technological changes to their established product lines. These changes tended to be competence enhancing—in other words, they encouraged the big three producers to reinforce their established organizational strengths (their core capabilities) and create higher barriers to entry, while many less successful competitors fell by the wayside, particularly during the nineteenth century. And, by the 1930s, ALCo, Baldwin, and Lima had so effectively eliminated Hughesian reverse salients in steam locomotive technology that steam locomotives were approaching the limits of their technological potential.²²

Diesels certainly represented a radical technological discontinuity, since they did not share any significant technology or components with steam locomotives and since their manufacture demanded vastly different organizational routines and competencies. Furthermore, since diesels offered a threefold increase in thermal efficiency over steam locomotives, there was little incentive for railroads to retain the old technology. Radical discontinuities tend to destroy competency, and this was certainly the case in the locomotive industry. The technological capabilities and organizational routines that had worked so well for so long in the steam locomotive industry were ill-suited to the vastly different requirements of steam locomotive production. Efforts to transfer basic design concepts (i.e., reliance on horsepower, rather than tractive effort; attempts to design a single high-power unit rather than linked low-power locomotives; and reliance on casting rather than welding) from one industry to the other created serious difficulties for steam locomotive producers in the diesel locomotive market.²³

At the same time, new entrants (particularly Electro-Motive) proved especially adept at managing this radical technological discontinuity and creating substantial first-mover advantages. Electro-Motive enjoyed the classic attacker’s advantage because that company had no financial, physical, or human resources invested in the core organizational capabilities of steam locomotive production.²⁴ Furthermore, Electro-Motive was free to explore a new set of performance characteristics and market applications (what Clayton M. Christensen and Richard S. Rosenbloom define as a value network) for diesel locomotive technology.²⁵ In other words, even though early diesels could not compete against steam locomotives under the set of performance criteria understood by the steam locomotive industry (high horsepower, low cost), they offered characteristics (operating efficiency, flexibility, relative cleanliness) that steam locomotives could not match. Even though two of the established steam locomotive builders had developed diesel locomotive technology before Electro-Motive entered the field, these efforts bore little fruit, in part because managers at ALCo and Baldwin saw quite clearly that these primitive machines were woefully inadequate, based on their traditional standards of evaluation.

The difficulty and sluggishness with which managers at established firms responded to the emergence of a new value network is often grounded in corporate managerial culture.²⁶ In any industry, if finances permit, physical facilities can be reconfigured, workers can be retrained, and new technologies can be developed or purchased. This process, difficult and expensive under the best of conditions, becomes much more complicated if a company is infused with a corporate managerial culture that places undue reliance on traditional product lines and production processes.²⁷ Although executives are rarely locked into a rigid corporate culture, and their companies are seldom locked into rigid technological trajectories, both corporate culture and technological trajectories constrain the ability of individuals and firms to respond to technological change.²⁸ Likewise, since corporate culture reduces flexibility on an individual level, technological paradigms often have the same effect on a corporate level.²⁹

Executives at ALCo and Baldwin, through lifelong training and experience, developed a corporate culture that was virtually inseparable from the custom craft production of steam locomotives. This corporate culture allowed ALCo and Baldwin to manage incremental changes in steam locomotive technology and thus dominate the steam locomotive industry and gain respect throughout corporate America as successful, reliable, and well-managed firms. Paradoxically, the corporate culture that contributed to this success had become ossified by the 1920s, with managers slow to respond to growing evidence that the steam locomotive industry was headed for extinction.³⁰ The very success of their corporate culture blinded executives at ALCo and Baldwin to the opportunities—and requirements—of radical technological change.³¹

At the same time, executives at Electro-Motive possessed a corporate culture that, while not inherently better than that of ALCo and Baldwin, nonetheless recognized the applicability of the diesel locomotive and fostered its development. In particular, executives at Electro-Motive, unlike their counterparts at ALCo, Baldwin, and Lima, realized that advances in diesel locomotive technology greatly increased the importance of advertising, marketing, and post-sales support services. The success of a company’s marketing efforts thus had as much effect on its profitability and long-term survival as did the applicability of its technology. And, while corporate culture was embodied in firm routines at the steam locomotive producers, it was largely independent of them at Electro-Motive—thus furthering the new entrant’s attacker’s advantage.³²

The locomotive industry offers ample evidence of corporate success and corporate failure. While a variety of factors, ranging from issues of technological transfer to instances of government action, influenced changing competitive patterns in the locomotive industry, endogenous factors proved to be the determinant of success or failure in the industry. As such, the industry offers examples of both the successes of individual initiative and of the limits of bureaucratic managerial control over the use of technology. In the locomotive industry, at least, deeply rooted corporate cultures ensured that the development of technology within the black box and corporate decisions regarding the applicability of that technology did not proceed predictably, or in a manner that could be consistently and effectively subsumed under a corporate hierarchical decision-making process.

The response of the American locomotive industry to technological change helps to illuminate the diverse responses of American business to the forces of change. The decline of established, respected, and profitable locomotive producers and the ascendancy and subsequent financial success of two new entrants into the field offers insights into the decline of established American industries and the rise of new industrial powers and new forms of economic endeavor.

I

Steam vs. Diesel: The Capabilities and Requirements of a Radically New Technology

Steam and diesel locomotives embodied vastly different technologies and, of even more concern to locomotive builders, these technological differences mandated radically different production and marketing techniques. In particular, even slight variations in operational requirements could require steam locomotive designs to be altered substantially. As a result, steam locomotive designs proliferated, and builders constructed steam locomotives in small batches, customized to suit the requirements of a particular railroad or operating district. Steam locomotives remained customized, purpose-built machines, and the necessity of tailoring locomotive designs to specific railroad requirements ensured that economies of scale were largely unobtainable and that customers would have the upper hand in the design process.

Diesels were more adaptable, however, ensuring that fewer designs could serve a wider variety of applications. This versatility, combined with high research and development costs, meant that diesels could be most efficiently produced in large batches to standardized designs. In other words, the shift from steam locomotive production to diesel locomotive manufacturing not only involved a radical transformation in technology; but that technological transformation also required companies and their managers to shift from customized, small-batch production techniques to the philosophy and methods of mass production.

Traditional Steam Locomotive Industry Practice

By the 1920s new steam locomotives were far larger than those built fifty years earlier, yet production techniques had changed little during the interval. Numerous builders produced small industrial steam locomotives; and some railroads built locomotives in their own shops. By the early years of the twentieth century, however, only three firms produced large mainline steam locomotives for independent railroad customers on a regular basis. Baldwin traditionally held a leadership role in the industry, although it traded market dominance with ALCo at regular intervals. Lima, a marginal producer, often accommodated the overflow from its two larger rivals. Between 1920 and 1928, ALCo enjoyed a 47 percent market share, followed by Baldwin, with 39 percent, and Lima, with 14 percent.¹

While the capital equipment sector traditionally exhibited close working relationships between suppliers and customers, this process was rarely more apparent than in the steam locomotive industry.² Railroad executives, particularly those in motive-power departments, demanded custom locomotive designs, and none more so than motive-power experts. Especially during the nineteenth century, this quest for uniqueness partly resulted from the vanity of motive-power officials who often believed that their designs were superior to those employed by their competitors. Railroads possessed differing operational characteristics (flat vs. mountainous terrain, freight vs. passenger traffic, impeccably maintained mainline vs. lightly built branchline), and this provided a more rational basis for custom orders. Despite design multiplicity, the steam locomotive builders, notably Baldwin, succeeded admirably in their efforts to standardize production techniques, primarily by using as many common components as possible for each locomotive design. The standardization of components did not extend to the locomotives themselves, however, and ALCo, Baldwin, and Lima produced a bewildering array of locomotive models. Baldwin, for example, offered 492 designs in 1915.³

ALCo, Baldwin, and Lima produced major locomotive components in separate buildings or areas, such as the tender shop, the frame shop, and the boiler shop. In these buildings, skilled craftsmen created parts and sub-assemblies to precise specifications, yet manufacturing tolerances were typically somewhat loose and parts often varied by as much as 1/100th of an inch. Skill and success in forging, casting, and machining ultimately determined the quality of locomotive components. At the end of the production process, locomotive components came together in the massive erecting shop, where detailed erecting cards specified how some six or seven thousand separate parts would fit together.⁴ In the erecting shop, skilled craftsmen built, rather than assembled, the locomotive piece by piece. In 1939, an observer wrote that one builder, Lima, had a pride in craftsmanship that has never been sacrificed to a conveyor belt..., at a time when the conveyor belt had already transformed the automobile industry, and many other industries as well.⁵

As was typical of many companies in the producer-oriented capital goods industry, the three steam locomotive producers possessed limited marketing capabilities. Top-level managers worked diligently to secure steam locomotive orders, relying on their close personal contacts with railroad motive-power officials. Builders generally solicited competitive bids from the railroads and were required to do so after passage of the Clayton Act in 1914.⁶

Steam locomotive builders offered little in the way of financing or post-sale support services. Baldwin, in particular, refused to provide financing for its steam locomotives, informing one customer that as you can appreciate, the Baldwin Locomotive Works as an equipment manufacturer is not in a position to undertake such financing for its own account.⁷ Instead, railroads typically sold equipment trust certificates to banks, insurance companies, and other major lenders in order to raise the funds necessary to purchase their steam locomotives. Since ALCo, Baldwin, and Lima custom built these locomotives to the unique specifications of particular railroads, their products could not be transferred easily to other railroads in the event of default. As such, lenders usually required an initial down payment—something they would not demand for purchases of standardized diesel locomotives.⁸

Builders did sell spare parts, which they cast or machined from wooden masters kept in a pattern vault. The multiplicity of steam locomotive designs ensured that many spare parts could not be kept in inventory, ready for immediate delivery. Additionally, since steam locomotives employed loose manufacturing tolerances, spare parts rarely fitted exactly. Once in railroad service, steam locomotives literally shook themselves apart, and the constant abrasion of metal parts against metal parts ensured that, even if the locomotive builders had achieved complete interchangeability in production, spare parts would never exactly fit the locomotive for which they were designed. Although some components, like headlights, bells, and cab assemblies, were essentially interchangeable, railroads still required highly skilled, highly paid, and effectively unionized railroad shop forces to fit spare parts to disabled steam locomotives.

Finally, since railroad employees understood steam locomotive operating and maintenance techniques, the steam locomotive builders had little incentive to offer training programs for railroad engineers or shop forces. At most, the builders supplied a representative who would ride with the engine crew of a new steam locomotive in order to evaluate its performance. As John Brown’s study of the Baldwin Locomotive Works points out, compared to other firms in the producer-oriented capital goods industry, builders such as Baldwin offered a surprising array of marketing